A machine learning-based recommendation system that suggests books using collaborative filtering and cosine similarity based on user preferences.

A full-stack video sharing platform where users can upload, explore, and download videos seamlessly. Built with Next.js 14, TypeScript, and Clerk for authentication, integrated with Cloudinary for video storage.

A supervised ML model that predicts the likelihood of diabetes based on health metrics like glucose, BMI, age, and blood pressure. Achieved 77% accuracy using scikit-learn models.

A Natural Language Processing project that detects fake and real news using TF-IDF vectorization and Linear SVM classifier.

An AI-powered system using Convolutional Neural Networks (CNNs) and MobileNetV2 to detect manipulated facial images (deepfakes). It features a Streamlit interface for image upload and real/fake detection.

An interactive global COVID-19 dashboard created in Tableau to visualize cases, recoveries, and deaths using multiple chart types including maps, trends, and bar charts.